Early maladaptive schemas mediating the relationship between perceptions of early parenting and depression and anxiety

Transcription

1 Rowan University Rowan Digital Works Theses and Dissertations Early maladaptive schemas mediating the relationship between perceptions of early parenting and depression and anxiety Monique Schatzman Rowan University Follow this and additional works at: Part of the Psychology Commons Recommended Citation Schatzman, Monique, "Early maladaptive schemas mediating the relationship between perceptions of early parenting and depression and anxiety" (2009). Theses and Dissertations This Thesis is brought to you for free and open access by Rowan Digital Works. It has been accepted for inclusion in Theses and Dissertations by an authorized administrator of Rowan Digital Works. For more information, please contact

2 EARLY MALADAPTIVE SCHEMAS MEDIATJING THE RELATIONSHIP BETWEEN PERCEPTIONS OF EARLY PARENTING AND DEPRESSION AND ANXIETY By Monique Schatzman A Thesis Submitted in partial fulfillment of the requirements of the Masters of Arts Degree Of The Graduate School at Rowan University May 4, 2009 Approved by. 175r. Jim A. Haugh Date ApprovedV Monique Schatzman

3 ABSTRACT Monique Schatzman EARLY MALADAPTIVE SCHEMAS MEDIATING THE RELATIONSHIP BETWEEN PERCEPTIONS OF EARLY PARENTING AND DEPRESSION AND ANXIETY 2007/08 Dr. James Haugh Master of Arts in Mental Health Counseling/Applied Psychology Abstract This study examined how EMS's mediate the relationship between perceptions of early parenting and depressive and anxious symptoms. The overall hypothesis was that EMS's based on the total score of YSQ-S3 would mediate the relationship between parental styles based on the YPI and depressive and anxious symptoms. Results indicated support for the mediating effects of EMS's in the relationship between perceptions of early parenting and depressive and anxious symptoms. EMS's mediated the overprotective, belittling, pessimistic/fearful, and punitive mother with depression. EMS's mediated the emotionally depriving, overprotective, belittling, pessimistic/fearful, controlling, and emotionally inhibited father with depression. EMS's also mediated the emotionally depriving, overprotective, emotionally inhibited, and conditional/narcissistic father with anxiety.

4 ACKNOWLEDGEMENTS Thank you to my thesis advisor, Dr. Haugh, and fellow student colleagues for helping me through this process. Thanks to my friends and family who provided me with support and encouragement. 11ii

5 TABLE OF CONTENTS Acknowledgements List of Tables and Figures CHAPTER I. Introduction What are Schemas? How Do Schemas Develop? How Do Schemas Influence Functioning and Psychopathology? Relations between Early Parenting and Negative Affect Relation between Schemas and Negative Affect Relation of Parenting to Negative Affect Relation between Parenting and Schemas Schemas as Mediators Limitations of the Previous Research and Hypotheses ii iv PAGE II. III. IV. Method Participants Design Procedure Materials Results Discussion References 34 11iii

6 LIST OF TABLES TABLE PAGE Table 1 Scales from the Young Schema Questionnaire- Short 3 (Young, 2005) 38 Table 2 Means and Standard Deviations for Father's Parental Styles and 40 Correlations between EMS's and Father's Parental Styles Table 3 Means and Standard Deviations for Mother's Parental Styles and 41 Correlations between EMS's and Mother's Parental Styles Table 4 Correlations between Father's Parental Styles and Depressive and 42 Anxious Symptoms Table 5 Correlations between Mother's Parental Styles and Depressive and 43 Anxious Symptoms Table 6 Means and Standard Deviations of Depressive and Anxious Symptoms 44 and Correlations between EMS's and Depressive and Anxious Symptoms Table 7 Step 1 of Mediation Model: Summary of Linear Regression Analyses for 45 Maternal and Paternal Styles of Parenting as Predictors of Total YSQ-S3 Table 8 Step 2 of Mediation Model: Summary of Linear Regression Analyses 46 for EMS's as Predictors of Depressive and Anxious Symptoms Table 9 Step 3 of Mediation Model: Summary of Linear Regression Analyses for 47 Maternal and Paternal Styles of Parenting as Predictors of Depressive Symptoms when Controlling for Anxiety Table 10 Step 3 of Mediation Model: Summary of Linear Regression Analyses 48 for Paternal Styles of Parenting as Predictors of Anxious Symptoms when Controlling for Depression Table 11 Step 4 of Mediation Model: Summary of Regression Analyses for the 49 Total YSQ S3 Score as a Mediator between Parental Styles and Depression Table 12 Step 4 of Mediation Model: Summary of Regression Analyses for the 50 Total YSQ S3 Score as a Mediator between Parental Styles and Anxiety FIGURE PAGE Figure 1 Mediation Model 51 iv

7 CHAPTER I Introduction A number of constructs have been linked to the development of depressive and anxious symptoms. One of the earliest factors found to predict depression and anxiety was perceptions of early parenting. Specifically, it was noted that low care and increased control from parents were found to predict the development of depression and anxiety in individuals (Harris & Curtin, 2002; McGinn, Cukor, & Sanderson, 2005; Shah & Waller, 2000). Another construct that has been linked to the development of depression and anxiety are early maladaptive schemas (EMS's). Specifically, it has been noted that the EMS's of defectiveness/shame, insufficient self-control, vulnerability, and incompetence/inferiority were predictive of depressive symptoms (Harris & Curtin, 2002; McGinn et al., 2005; Shah & Waller, 2000). The EMS's associated with the Overvigilance/Inhibition Domain were predictive of anxious symptoms (McGinn et al., 2005). Although both EMS's and perceptions of early parenting have been linked to depression and anxiety, little research has been conducted to examine how EMSs and perceptions might be related to one another in predicting depression and anxiety. Theoretically, one way these constructs might be related is that parenting causes the development of schemas which, in turn, cause depression and anxiety. In order to test this model, a mediational analysis must be done. A mediation analysis gives a more functional understanding of the relationship among the variable discussed. Examining these constructs within a mediational model attempts to explore more specifically the links between perceptions of parenting styles, the mediating effects of cognitive styles such as EMS' s, and the development of depression and anxiety. The goal of this

8 study is to examine how perceptions of early parenting and early maladaptive schemas are related to the development of depression and anxiety in a mediation model. What are Schemas? Two of the most prominent schema theorists are Aaron Beck and Jeffery Young. Beck (1976) defines schemas as being deep cognitive structures that enable an individual to interpret his or her experiences in a meaningful way. Similarly, Young (1990) defines schemas as enduring themes regarding oneself and one's relationship with others that are used every day as structures for organizing information and making meaning about the world. According to both Young and Beck, we have multiple schemas. Beck (1976), states that there are four different types of schemas, which are the views of the self, others, the world, and the future. In contrast, Young (1994) proposed that there are 18 EMS's, each of which is defined in Table 1. Young further hypothesized that these EMS's could be organized into five domains. The five domains are Disconnection and Rejection, Impaired Autonomy and Performance, Impaired Limits, Overvigilance and Inhibition, and Other Directedness. These domains are related to the basic emotional needs of a child and the 18 EMS's are grouped in these five broad domains on the basis of which core needs the schema is related to. How Do Schemas Develop? Beck's (1976) descriptions of the development of EMS' s are very similar to Young's description since much of Young's work was integrated from the work of Beck. Young proposed that these schemas develop in childhood from interplay between the child's innate temperament, and the child's ongoing damaging experiences with parents, siblings, or peers. Because they begin early in life, schemas become familiar and thus comfortable. Young (1990) proposed that these cognitive structures continue to develop over time

9 How Do Schemas Influence Functioning and Psychopathology? Whether the positive or negative aspect of a schema is activated depends on many different factors. According to cognitive theory, problems arise both when negative schema are too strong or activated too often and when balancing or positive schema are too weak or activated too infrequently (Beck, 1976). Schemas also may remain dormant until they are activated by situations relevant to that particular schema. Young (1990) proposed that EMS's are present in normal populations, but may become negative in symptomatic persons. We distort our view of the events in our lives in order to maintain the validity of our schemas. According to Beck's theory of the etiology of depression, negative schemas are acquired during childhood and adolescence. Examples of how these schemas are acquired are through a loss of a parent, rejection of peers, criticism from teachers or parents, the depressive attitude of a parent and other negative events. When the person with such schemas encounter a situation that resembles in some way, even remotely, the conditions in which the original schema was learnt, the negative schemas of the person are activated. These schemas are also believed to be formed in response to life experiences over a lifetime (Young, 1999). Beck and Young also believed that these schemas are related to pathology. Young (1994) stated that schemas represent coreprocessing domains that may manifest into interpersonal and relational malfunctioning or into the development of specific pathology in an individual. As stated before by Beck's theory of the etiology of depression, a person has negative schemas and cognitive biases in their mind. A cognitive bias is a view of the world. Depressed people, according to this theory, have views such as "I never do a good job." A negative schema helps give rise to the cognitive bias, and the cognitive bias helps fuel the negative schema. Also, Beck (1976) proposed that depressed people

10 often have the following cognitive biases: arbitrary inference, selective abstraction, overgeneralization, magnification and minimization. These cognitive biases are quick to make negative, generalized, and personal inferences of the self, thus fueling the negative schema. Relations between Early Parenting and Negative Affect Depression has been linked to exposure to prenatal stress, abuse and neglect, early trauma, and loss of a parent (Goodman, 2003). Infants and young children who are exposed to negative styles of parenting may develop insufficient interpersonal skills and emotion regulation creating a higher susceptibility to having depression in adulthood. Parents who are coercive or controlling can contribute to a child developing a sense of helplessness. The child can view themselves as having little control over outcomes which my carry over into their adulthood relating to the development of depression (Racusin & Kaslow, 1991 as cited in Goodman, 2003). Vulnerability to depression can also occur when there is abnormal family functioning, which contributes to the development of negative patterns of coping, beliefs, and interpersonal styles (Goodman, 2003). Similar to depression, anxiety has also been linked to negative styles of parenting. Those with anxiety, on the other hand, tend to react to potential threat by engaging in greater cognitive avoidance and/or emotional avoidance instead of attending to additional environmental information indicating that a potential threat is actually dissipating (Riskind & Williams, 2005). Relation-between Schemas and Negative Affect Cognitive theory hypothesizes that one of the cognitive constructs that should be related to affect and affective disorders are schemas (Beck & Perkins, 2001). This hypothesis has been studied most in relation to depression and anxiety. With regard to depression, the current evidence seems to suggest that there is a relationship between schemas and depression. For

11 example, McClain and Abramson (1995) studied the relationship between schemas and depression in college students. Results indicated that there was an increased vulnerability to depression when students had less positive schemas. There are some studies that have also examined the relationship between schemas and the development of anxiety. Pinto- Gouveia, Castilho, Galhardo, and Cunha (2006) compared EMS's between a group of patients with social phobia, other anxiety disorders, and a group of non-psychiatric controls. Results indicated that both clinical groups scored significantly higher than the general population group in most of the subscales of the schema questionnaire. The subscales associated with significantly higher scores were the EMS's: emotional deprivation, guilt/failure, social undesirability/defectiveness, mistrust/abuse, social isolation/alienation, dependence, abandonment, subjugation, and shame. Calvete et al. (2005) investigated how EMS's can predict both depressive and anxious symptoms in a non-clinical sample. Results indicated that defectiveness/shame, failure, and selfsacrifice were the only significant EMS's to predict depressive symptoms. Also, abandonment, failure, and subjugation were the only significant EMS's to predict anxious symptoms. Stopa, Thorne, Waters and Preston (2002) further tested the ability of EMS's to predict depressive and anxious symptoms except in a clinical sample. Results indicated that abandonment, defectiveness, subjugation, and self-sacrifice were the EMS's that significantly predicted depressive symptoms. Also, results indicated that unrelenting standards was the only EMS that significantly predicted anxious symptoms. Welbumn, Coristine, Dagg, Pontefract, and Jordan (2002) also examined the ability of EMS's to predict depression and anxiety in a clinical sample. Results indicated that there was a significant difference in EMS' s predicting the development of psychiatric symptoms including

12 the symptoms of depression and anxiety. Specifically abandonment, insufficient self-control, and dependency EMS's were significant predictors of depressive symptoms. Also, abandonment, vulnerability to harm, failure, self-sacrifice and emotional inhibition EMS's were significant predictors of anxious symptoms. Similarly, Glaser et al. (2002) also examined EMS's importance in the development of ability to predict depression and anxiety in adulthood. This study also examined which specific schemas are related to the components of depressive and anxious symptoms. Results from this study found that the schemas of abandonment/instability and vulnerability to danger/random events were significant predictors of depressive and anxious symptoms. Overall results indicated that all of the Early Maladaptive Schemas Questionnaire-SF subscales, with the exception of entitlement/self-centeredness, were significantly correlated with the depression and anxiety scales. In summary, results from previous literature investigating the ability of specific EMS's to predict depressive and anxious symptoms have been inconsistent. The inconsistency in the results can be because each study used different methodology such as how anxious and depressive symptoms were measured, different procedures of scoring, and different types of samples. On the other hand, these studies do indicate a pattern to support that EMS's do account for a significant portion of the variances in both depressive and anxious symptoms. In addition, these results indicated that different EMS's were significant predictors of either depression or anxiety. Relation of Parenting to Negative Affect The development of anxiety and depression has also been investigated by studying the influence of the family on the development of anxious and depressive symptoms. There are a 6

13 number of ways in which familial influence has been defined in these studies. One way familial influence has been examined is by studying the global perception of one's family climate. For example, Oliver, Klocek, and Wells (1995) examined the relationship between family climate and depressive and anxious symptoms. A sample of undergraduate students completed measures on their perceptions of their relationships with their parents and the climate in their family of socialization, subjects' attention on the self, and current symptoms of depression and anxiety. Results indicated that the participants' perceptions were significantly related to cognitive vulnerability to depression and most likely to anxiety when parents do not support participants' psychological autonomy. There was also a lesser degree of current symptoms of state anxiety associated with cognitive vulnerability. Furthermore, increased closeness to one's mother may be accompanied by an increased awareness of both external and internal aspects of the self. A second way to examine familial influence is by specifically examining thematernal and paternal parental characteristics. Baumrind (1968) examined the relationship between both maternal and paternal parenting styles with depressive and anxious symptoms in an adolescent non-clinical sample. Results indicated significant positive correlations between negative parenting styles and depressive and anxious symptoms. There were also significant negative correlations between adaptive parenting styles and depressive and anxious symptoms. Parker (1979b & 1981) also examined parental characteristics in relation to depression and anxiety; however, they utilized a clinical sample. In the investigation by Parker (1979b), results indicated significant positive correlations between negative parenting styles and anxious symptoms. Results also indicated significant negative correlations between adaptive parenting styles and both depressive and anxious symptoms. The results of Parker (1981) also indicated

14 significant positive correlations between negative parenting with depressive symptoms and significant negative correlations between adaptive parenting styles with depressive symptoms. A third way of examining familial influence is by focusing on how individuals perceive the different characteristics of how their parents raised and disciplined them. For example, perceived parental rearing practices measured by "Affection" and "Control" in depressed and anxious patients were reviewed in a meta-analysis by Gerlsma, Emmelkamp, and Arrindell (1990). The terms "Affection" and "Control" relate to parental characteristics of care and overprotection as measured by the PBI. In regards to anxious symptoms, results indicated that various types of phobic disorder were related to a parental rearing style of less "Affection" and more "Control" as compared to healthy controls. However, findings for depression seemed less consistent, but it was found that the largest effect sizes were found for less "Affection" and more "Control." A final way of examining familial influences is by observing the link between attachment style and parental behaviors. For example, Kenny and Sirin (2006) investigated the characteristics of parental attachment in young adults. The role of the attachment figure was explored as a secure base by assessing whether financial support from parents and frequency of communication are associated with perceptions of parental attachment and young adults' selfreports of self-worth and depressive symptoms. Results indicated that emerging adults' perceptions of parental attachment were associated positively with perceived self-worth and negatively with levels of depressive symptoms. In other words, if a person perceives their parental attachment relationship negatively, then they are more likely to be experiencing depressive symptoms.

15 Similar to Kenny and Sirin (2006), Siqueland, Kendall, and Steinberg (1996) examined parental attachment differences between families with an individual diagnosed with an anxiety disorder and families without an individual diagnosed with an anxiety disorder. The focus on this study was on anxiety rather than depression. Results indicated that parents of individuals with an anxiety disorder were rated as less accepting and more controlling. Based on the results of these studies on the relationship between perceptions of early parenting and depressive and anxious symptoms, negative parenting styles, behaviors such as controlling/overprotection, how individuals perceived their parents' parenting styles, and negative attachment styles have all been significantly associated with symptoms of depression and anxiety. However, since the correlations of the previous research were modest, it is possible that other factors may influence the relationship between perceptions of early parenting and the development of depression and anxiety. Relation between Parenting and Schemas Perceptions of early parenting have also been related to schema development. Bowlby (1979) stated that adverse experiences during childhood play a large part in causing cognitive disturbance. There are three distinct categories of experiences that tend to be extremely influential in affecting thought, feeling, and behavior: those that parents wish their children not to know about, those in which parents have treated children in ways the children find too unbearable to think about, and those in which children have done, or perhaps thought, things about which they feel unbearably guilty or ashamed. Most children overlook many deficiencies by their parents because they would rather see their parents in a favorable light (Sheffield, Waller, Emanuelli, Murray, & Meyer, 2006). These experiences that occur in childhood may result into the development of EMSs as stated by Young (1990). Young (1990) believed that

16 EMSs develop during childhood through relationships with significant caretakers. The development of these EMS's are selectively filtered and the schemas are elaborated throughout an individual's life. Andersson and Perris (2000) examined a different way of identifying schemas. In this study the possible relationship between experiences of dysfunctional parenting and the development of dysfunctional assumptions of the self and others was examined in a non-clinical sample. Results indicated that experiences of dysfunctional parenting were positively correlated with dysfunctional working models of the self and others. Even though this study did not specifically examine schemas, these dysfunctional working models can be related to the development of EMS's. Therefore it can be assumed that perceptions of early parenting are related to the development of EMS's from this study. Another way that early parenting and schemas have been studied is by Sheffield et al. (2006) who specifically examined the links between parenting and EMS's based on Young's theory. Based on a sample of undergraduate and postgraduate students, results indicated that parenting has a general effect on negative core beliefs. Unlike the previous study, Sheffield et al. (2006) measured early parenting and schemas using the Young Parenting Inventory (YPI-R) instead of the Egna Minnen av Barndoms Uppfostran (EMBU). This study also used the Young Schema Questionnaire- Short Form (YSQ-SF) to measure EMS's, whereas Andersson and Perris (2000) used the Dysfunctional Attitudes Scale (DAS-S) and the Dysfunctional Working Models- Scale (DWM-S). Even though Sheffield et al. (2006) found a significant relationship between early parenting and negative core beliefs, it was not able to be concluded that all the schemas were directly related with all the scales in perceived parenting experiences. 10

17 Beck's cognitive model of depression (1967) explicitly attributes the development of negative schematic and cognitions to a critical, disapproving parent (as cited in Shah & Waller, 2000). Poor perceived parental care and overprotection before the age of 16 are associated with neurotic and reactive depressive disorders. There is evidence of a role for interpersonal factors, particularly the impact of poor parental care on the formation of deficient social bonds. Shame, coping resources, self-esteem, and a collection of personality vulnerability factors (neuroticism, dysfunctional attitudes, self-esteem, and external locus of control) are examples of factors that can form from poor parental care (Shah & Waller, 2000). Schemas as Mediators These EMS's may also be effective cognitive mediators. A mediator is also known as an intervening or process variable that explains how external physical events take on internal psychological significance (Baron & Kenny, 1986). Bowlby (1988) and Young (1990) propose that internal working models or schemas may mediate the relationship between early experiences and the development of pathology. Some studies have attempted to examine schemas as a mediator between perceptions of early parenting and types of psychopathology, specifically depression and anxiety. For example, Shah and Waller (2000) examined the role of core beliefs or schemas as a mediator in the relationship between recalled parenting in childhood and major depression in adults. Using a clinical sample of adult, depressed outpatients, results indicated that only five core beliefs partially mediated the relationship between maternal bonding and paternal overprotection and levels of depression: dependence/incompetence, emotional inhibition, failure to achieve, unrelenting standards, and vulnerability to harm. It was concluded in this study that 11

18 specifically poor care and high control parenting can create maladaptive cognitive schema which would make an individual vulnerable to depression. Harris and Curtin (2002) also examined the mediating relationship between retrospective reports of parenting, EMS's, and symptoms of depression; however, it was observed in a sample of undergraduate students rather than a clinical sample. Results indicated that the schemas of defectiveness/shame, insufficient self-control, and vulnerability to harm partially mediated the relationship between perceptions of early parenting and depressive symptoms. The schema of incompetence/inferiority was also found to be a partial mediator; however, it only mediated the relationship with the care scale of the PBI and not the overprotection scale. McGinn, Cukor, and Sanderson (2005) also examined the mediating relationship of cognitive styles between early experiences with low care, increased control, abuse and neglect, and psychopathology in a clinical sample. However, this study expanded from previous research by attempting to examine both depression and anxiety and used another scale for measuring negative parenting along with the PBI called the Childhood Trauma Questionnaire (CTQ). Since there was no significant correlation between the PBI scales with the BAI and BDI, a mediation analysis could not be performed with the PBI. Also, total scores from the CTQ were highly correlated with only depression not anxiety; therefore, a mediation analysis was able to be performed between the CTQ, cognitive style, and depression. Results indicated that three domains out of the five from the YSQ mediated the relationship between childhood abuse and depression: disconnection/rejection, impaired autonomy/performance, and impaired limits. The schemas associated with these three domains are abandonment/instability, mistrust/abuse, emotional deprivation, defectiveness/shame, social isolation/alienation, failure to achieve, 12

19 vulnerability to harm/illness, functional dependence/incompetence, enmeshment, entitlement, and insufficient self-control. Limitations of the Previous Research and Hypotheses Together, these studies provide some evidence for the role of cognitive mediation in the relationship between perceptions of early parenting and the development of psychopathology. However, there are many limitations to these studies which the present study attempts to correct. One limitation of the previous research is that only depression was studied in the relationship between perceptions of early parenting and schemas. Although McGinn et al. (2005) attempted to analyze anxiety along with depression; there was no significant correlation between the measure of parenting and the measure of anxiety so the mediation analysis could not be completed. Therefore, there has still yet to be any clear results stating that there could be a mediating relationship between perceptions of early parenting, early maladaptive schemas, and both anxiety and depression. The second limitation is the inconsistent support in the relationship of depressive and anxious symptoms for the cognitive content-specificity hypothesis. Previous researchers have failed to separate the symptoms of both depression and anxiety. It is important to examine both anxiety and depression because their symptoms overlap with one another. Also, when studying both, the shared variance can be eliminated, so there is a clearer prediction of what is related to depression versus anxiety. The third limitation is that the previous research does not examine specific early experiences relating to specific schemas. This study attempts to see which specific schemas relate to either depression, anxiety, or both. This would make it more clear what specific characteristics relate to depression and anxiety especially in the treatment of these symptoms. It 13

20 is also important to examine the specific parenting styles related to specific EMS's in order to be able to better predict what parental characteristics can develop into which EMS's. The fourth limitation is that majority of the studies use the PBI to measure characteristics and perceptions of parenting. The PBI only measures care and overprotection of the mother and the father which is a broad overview. However, the Young Parenting Inventory measures the origins of each schema based on how participants characterize their childhoods with their parents. Since parenting goes deeper than just care and overprotection, it is important to research other areas of parenting that might explain schema development and psychopathology. The final limitation is the few studies which investigate the mediating role of EMS's in the relationship between perceptions of early parenting and depression and anxiety. The current study extends from the previous literature by changing upon the methods utilized previously and controlling for both depression and anxiety when testing the mediation model. The present study attempts to improve upon the limitations of existing research as previously stated. Therefore, the purpose of this study is to understand the factors predict depression and anxiety. First it is hypothesized that negative perceptions of early parenting affect the development of depression and anxiety in adulthood. Second, it is hypothesized that negative perceptions of early parenting also contribute to the development of early maladaptive schemas. Third, it is hypothesized that early maladaptive schemas are related to depression and anxiety. Fourth, a-mediation model is proposed; specifically that early maladaptive schemas mediate the relationship between perceptions of early parenting and depression and anxiety. 14

21 CHAPTER II Method Participants Participants were 102 undergraduate students located in a public, mid-sized, northeastern university. Individuals were recruited from Introduction to Psychology classes and received course credit for being a participant in the study. There were 49 (48%) males and 53 (52%) females who ranged from ages of 17 to 32 years old with a mean age of 19.9 years. The class ranks of the students were 30.4% freshman, 26.5% sophomores, 26.5% juniors, and 16.7% seniors. Fifty-one percent of the participants were single, 3.9% were engaged, 42.2% were single but in a committed relationship, and 2.9% reported their relationship status as other. In order to examine the generalizability of the current sample to the general university sample, the Fall 2006 enrollment data from Rowan University's Office of Institutional Research and Planning (2006) was utilized. According to that data, there were 78.5% Caucasian, 8.8% African-American, 6.7% Hispanic, 3.0% Asian, and 0.4% American Indian/Native Alaskan, and 2.6% of students did not report. This was consistent in this study's sample by 87.3% of Caucasian, 4.9% of African-American, 2.0% of Hispanic, 2.0% of Asian, and 3.9% reported as being other. This study also included 49% of people living in an on-campus dorm, 29.4% living off campus in a apartment or house, 20.6% live at home with parents or primary caregivers, and 1% reported living as other. The primary caregivers of the students were reported as having 82.4% with two biological parents, 10.8% having one biological parent, 4.9% having one biological parent and one step-parent, 1.0% having adoptive parents, and 1.0% having other. Based on this data, the sample appears to be an adequate and representative sample of the population. 15

22 Design This correlational study used a cross-sectional design. The independent variables that were under investigation were EMSs and perceptions of early parenting, and the dependent variables were the ratings of depressive and anxious symptoms. Figure 1 shows the representation of the meditational model that was examined in this study. This model examines whether EMS's mediate the relationship between perceptions of early parenting and symptoms of depression and anxiety. Procedure This project was approved by the university's Institutional Review Board and conformed to the ethical codes of the American Psychological Association. Prior to participating in the study, participants reviewed and signed an informed consent form that was reviewed and discussed with them by one of the experimenters. Participants were also debriefed following their participation in the study. Each participant completed four self-report questionnaires which were administered in groups of 2-30 people that took approximately 60 minutes to complete. The YSQ-S3, BDI-II, BAI, and the YPI were given out first followed by a brief demographic questionnaire that asked participants about their age, sex, ethnicity, and class rank. Participants also reported if they lived on campus or at home and which primary caregiver was present in the home during the first 18 years of their lives. There living situation was described as living with either two biological parents, one biological parent, one step parent and one biological parent, adoptive parents, or other. Materials Young Schema Questionnaire-Short 3 (YSQ-S3; Young, 2005). The YSQ-S3 is a 90- item, self-report questionnaire which is an extension from the Young Schema Questionnaire- 16

23 Short Form (Young & Brown, 1994). Each item uses a 6 point Likert scale ranging from 1 ("completely untrue of me") to 6 ("describes me perfectly"). The difference between the YSQ- SF and YSQ-S3 is that the YSQ-SF measures 15 out of the 18 EMS's proposed by Young (1994), whereas, the YSQ-S3 measures all 18 of Young's EMS's which are briefly described in Table 1. Each scale consists of 5 items, and the scores for each EMS are calculated by summing the items for each EMS. The higher score on a scale indicates a greater presence of that particular EMS. Since there are currently no studies which have examined the psychometric properties of the YSQ-S3, the psychometric properties of YSQ-SF are reported. The YSQ-SF has shown good internal consistency (i.e., a coefficients ranging from.74 to.89), and correlations with positive and negative affectivity, depression, and self-esteem in clinical samples, suggesting it has good predictive validity (Calvete et al., 2005). The YSQ-SF is also considered to be a valid measure of EMSs. The construct validity of the YSQ-SF was investigated by Welburn et al. (2002) and through a factor analysis on a clinical sample found that there was a strong correspondence between the theoretical subscale structure of all 15 EMSs and the resulting analysis. Out of the 74 items, 70 loaded exactly with the theoretical structure of the instrument, suggesting there is strong construct validity for all scales on the YSQ-SF. Welburn et al. (2002) also observed the criterion validity of the YSQ-SF by examining the relationship between the 15 EMSs and different types of psychological distress, such as depression and anxiety. Results indicated that different types of psychological distress were significantly predicted by EMSs. The criterion validity of the YSQ-SF was further investigated by Glaser et al. (2002) by examining the relationship between the 15 EMSs and various measures 17

24 of psychological symptoms in a clinical sample. Glaser et al. also found strong criterion validity because specific EMSs have been shown to predict high levels of affective disturbances. Beck Depression Inventory-II (BDI-II; Beck, Steer, & Brown, 1995). The BDI-II is a 21- item, self-report measure designed to assess symptoms of depression along affective, cognitive, motivational, and physiological dimensions. Each item is based on a 4 point Likert scale ranging from, 0 ("absence of symptom") to 3 ("severe symptom"), and total scores range from Dozois, Dobson, and Ahnberg (1998) examined the psychometric properties of the BDI-II by comparing it to the BDI. Results indicated that the BDI-II was a stronger instrument than the original BDI based on factor structure. The BDI-II was also found to be internally consistency with a coefficient alpha of.91. Beck Anxiety Inventory (BAI; Beck, Steer, & Brown, 1993). The BAI is a 21-item, selfreport measure of common symptoms of anxiety and their severity. Each item describes a symptom of anxiety and has response based on a 4 point Likert scale ranging from, 0 ("absence of symptom") to 3 ("severe symptom"), and total scores range from Total scores are calculated by summing the responses for each of the 21 items. The BAI has proven to be reliable and valid. For example, in a study of Caucasian-American and Latino college students the BAI had good reliability with an internal consistency coefficient that exceeded.82 (Contreras, Fernandez, Malcarne, Ingram, & Ruiz Vaccarino, 2004). Young Parenting Inventory (YPI; Young, 1999). The YPI is a self-report measure consisting of 72 items that individuals might use to describe their parents. Each item is rated on a two six-point Likert scale ranging from 1 ("completely untrue") to 6 ("describes him/her perfectly"), with the exception of the emotionally deprivation scale, which is reversed scored. 18

25 Higher scores indicate a perception that the parent behaved in ways that were more likely to generate the related core beliefs. There are three different ways in which the YPI can be scored. The first way of scoring the YPI is to sum all the items on each scale where high scores of 5 and 6 are converted to 1 and scores of 1 through 4 are converted to 0 (Young, 1999). For example, the two-letter code for items 1 through 5 is marked with *ed which represents the emotional deprivation scale. Since the emotional deprivation schema is reversed scored, the items that are scored 1 and 2 are summed for the total score for that scale. The second way to score the YPI is to sum all the items for each scale the same way as Young (1999) except adding each item individually by the by the number represented. For example, if scores for the abandonment schema represented by *ab have scores of 6, 3, 5, and 2, those numbers would be summed for a total score of 16 for the abandonment scale. The final way of scoring the YPI, which was adopted in this study, was described by Sheffield et al. (2006). According to Sheffield et al. (2006), the YPI consisted of nine factors that were common to both parents the data supported it from the factor analysis. The emotionally depriving parenting scale reflects a pattern of parenting that deprives the child of emotional nurturing. The overprotective parenting scale reflects a pattern of parenting that overprotects the child. The belittling parenting scale reflects behavior in the parent that belittles the child, leaving the child feeling defective. The perfectionist parenting scale reflects the parents' expectations for themselves as well as for their child. The pessimistic/fearful parenting scale reflects anxious, fearful traits in the parent. The controlling parenting scale reflects a pattern of parenting that controls or inhibits the child's independence. The emotionally inhibited parenting scale reflects the parent's ability to share their feelings with their child. The punitive parenting scale reflects a 19

26 pattern of parenting that is punitive of the child's mistakes. The conditional/narcissistic parenting scale reflects behavior by the parent that implies that positive regard for the child is conditional on the child's success. To calculate the scores, all the items for each of the nine scales were summed. Sheffield et al. (2006) also investigated the psychometric validation of the YPI by having a large non-clinical student sample (N = 422) complete the YPI and a subset also completed the YSQ-SF. All nine scales had good test-retest reliability and adequate internal consistency. The YPI also had adequate construct validity. Both maternal and paternal YPI-R scales were commonly associated with a number of YSQ-SF scales. All YPI-R scales correlated significantly with some YSQ-SF scales with the exception of Perfectionist mothers and fathers and pessimistic/fearful fathers. Also, the YPI-R scales were related to all 15 negative core beliefs as measured by the YSQ-SF except for the self-sacrifice scale. The significant associations between the YPI-R scales and negative core beliefs as measured by the YSQ-SF partially supported the construct validity of the YPI. The psychometric validity of the YPI is at an acceptable level that can potentially be used to explain part of the variance in negative core beliefs. 20

27 CHAPTER III Results The first goal of this study was to examine the relationships between early parenting, schemas, and depression and anxiety by using the Pearson-r correlation. Means and standard deviations for the father's parental styles and correlations between EMS's and the father's parental styles are displayed in Table 2. Fourteen out of the 18 correlations between EMS's and the emotionally depriving father were statistically significant, with the significant coefficients ranging from r =.21 to r =.48. Fifteen out of the 18 correlations between EMS's and the overprotective father were statistically significant, with the significant coefficients ranging from r =.20 to r =.46. Thirteen out of the 18 correlations between EMS's and the belittling father were statistically significant, with the significant coefficients ranging from r.21 to r.40. None of the correlations between EMS's and the perfectionist father were statistically significant. Twelve of the 18 correlations between EMS's and the pessimistic/fearful father were statistically significant, with the significant coefficients ranging from r=.21 to r =.39. Fourteen out of the 18 correlations between EMS's and the controlling father were statistically significant, with the significant coefficients ranging from r -.23 to r Thirteen of the 18 correlations between EMS's and the emotionally inhibited father were statistically significant, with the significant coefficients ranging from r =.20 to r=.39. Eleven of the 18 correlations between EMS's and the punitive father were statistically significant, with the significant coefficients ranging from r =.21 to r =.38. Five of the 18 correlations between EMS's and the conditional/narcissistic father were statistically significant, with the significant coefficients ranging from r =.23 to r Eight of the 9 correlations between the total YSQ-S3 Total Score and the different parental styles for the father were statistically significant, with the significant 21

28 coefficients ranging from r =.25 to r =.46. Overall results from Table 2 indicate that majority of the YPI scales for the fathers were related to the majority of EMS's. However, the perfectionist father scale was not significantly related to any of the EMS's. Means and standard deviations for the mother's parental styles and correlations between EMS's and the mother's parental styles are displayed in Table 3. Five out of the 18 correlations between EMS's and the emotionally depriving mother were statistically significant, with the significant coefficients ranging from r = -.21 to r = All of the significant correlations were in the negative direction. Eight out of the 18 correlations between EMS's and the overprotective mother were statistically significant, with the significant coefficients ranging from r=.23 to r.35. Seven out of the 18 correlations between EMS's and the belittling mother were statistically significant, with the significant coefficients ranging from r =.20 to r.28. Two of the 18 correlations between EMS's and the perfectionist mother were statistically significant in the positive direction. Fourteen of the 18 correlations between EMS's and the pessimistic/fearful mother were statistically significant, with the significant coefficients ranging from r =.20 to r-.50. Sixteen out of the 18 correlations between EMS's and the controlling mother were statistically significant, with the significant coefficients ranging from r =.22 to r =.56. Four of the 18 correlations between EMS's and the emotionally inhibited mother were statistically significant, with the significant coefficients ranging from r=.21 to r=.32. Nine of the 18 correlations between EMS's and the punitive mother were statistically significant, with the significant coefficients ranging from r =.21 to r =.40. Four of the 18 correlations between EMS' s and the conditional/narcissistic mother were statistically significant, with the significant coefficients ranging from r-=.21 to r =.31. Five of the 9 correlations between the total YSQ-S3 22

29 Total Score and the different parental styles for the mother were statistically significant, with the significant coefficients ranging from r =.23 to r =.54. Overall results from Table 3 were very similar to the YPI scales for the father, except the unrelenting standards schema and the self-sacrificing schema were significantly related to the perfectionist mother. However, there were less significant correlations between EMS's and YPI scales for the mother overall in comparison to YPI scales for the father. Another difference between the father and mother scales was that the father's overprotective style of parenting was related to the most EMS's, whereas the mother's controlling style of parenting was related to the most EMS's. Correlations between the father's parental styles and depressive and anxious symptoms are displayed in Table 4. Six of the 9 correlations between depressive symptoms and the father's parental styles were statistically significant, with the significant coefficients ranging from r-.28 to r =.40. Five of the 9 correlations between anxious symptoms and the father's parental styles were statistically significant, with the significant coefficients ranging from r-.21 to r=.43. Correlations between the mother's parental styles and depressive and anxious symptoms are displayed in Table 5. Six of the 9 correlations between depressive symptoms and parental styles were statistically significant, with the significant coefficients ranging from r-.21 to r All of the significant correlations were in the positive direction except for the relationship between emotionally depriving mother and depression, which was negative. None of the correlations between anxious symptoms and mother's parental styles were statistically significant. Overall results from Table 4 and 5 indicated that there were some differences between the relationship between the father and mother's parental styles with depressive and anxious 23

30 symptoms. For example, the emotionally depriving father, belittling father, emotionally inhibited father, and conditional/narcissistic father were significantly correlated with anxious symptoms; whereas, none of the mother's parental scales were significantly correlated with anxious symptoms. Also, the emotionally inhibited father was significantly correlated with depressive symptoms; whereas, it was not for the mother. The punitive mother was significantly correlated with depression; whereas, it was not for the father. Means and standard deviations of depressive and anxious symptoms and correlations between EMS's and depressive and anxious symptoms are displayed in Table 6. All of the correlations between EMS's and depressive symptoms were statistically significant except for the unrelenting standards schema, with the significant coefficients ranging from r=.26 to r =.62. The correlation between the total YSQ-S3 score and depressive symptoms was statistically significant (r =.69). All of the correlations between EMS's and anxious symptoms were statistically significant except for the unrelenting standards schema and punitiveness schema, with the significant coefficients ranging from r =.21 to r The correlation between the total YSQ-S3 score and anxious symptoms was statistically significant (r =.50). Overall the results displayed in Table 6 indicate that all the EMS's were significantly correlated with depressive and anxious symptoms, with the exception of the punitive schema which was not correlated with anxious symptoms and the unrelenting standards schema which was not associated with both depressive or anxious symptoms. Mediation Model The second goal of the study is to test whether or not EMS's mediate the relationship between perceptions of early parenting and depressive and anxious symptoms when controlling for both. To demonstrate the mediating effect of the EMS's as stated by Baron and Kenny 24